Skip to content
View SayedNadim's full-sized avatar
🎯
Nonchalant
🎯
Nonchalant
39 followers · 43 following

Achievements

Achievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLOAchievement: StarstruckAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLOAchievement: StarstruckAchievement: Arctic Code Vault Contributor

Highlights

Block or report SayedNadim

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
sayednadim/README.md

S M Nadim Uddin

Computer Vision Researcher · M.Sc.

I work on perception systems, from the research problem through to something that runs on real hardware. My work covers 2D and 3D image processing, generative modeling, and multi-sensor data, much of it aimed at in-cabin and automotive settings. I like questioning assumptions that get treated as settled, and I'm open to founding and early-stage work.

Based in Seoul, South Korea. Originally from Bangladesh.

Website LinkedIn Google Scholar ResearchGate

What I do

At Deep In Sight, my current role, I work in the R&D group on driver and in-cabin monitoring. That means object detection, semantic and instance segmentation, monocular depth estimation, and the model compression and pipeline integration needed to get these running in production rather than just on a benchmark.

Before that I spent two years at DeltaX.ai in the Automotive Perception Group, moving from researcher to project lead to group lead. The work spanned driver and occupancy monitoring, depth estimation, 3D reconstruction, body keypoints, and sensor fusion across RGB, IR, LiDAR, IMU, and RADAR, along with the detection, segmentation, and planning around them.

Current focus

  • In-cabin and driver monitoring, plus monocular depth estimation
  • Detection and segmentation, semantic and instance
  • Model compression and deployment for edge devices
  • 3D reconstruction and multi-sensor fusion

I work primarily in PyTorch with some C++, and also use TensorFlow, ONNX, TFLite, and ROS.

Open to

  • Founding or joining early-stage teams building serious vision and perception products
  • Technical leadership where research and engineering meet
  • Research collaborations and open-source work worth doing

Selected publications

  • Multi-Scale Attention-Guided Non-Local Network for HDR Image Reconstruction. Sensors, 2022.
  • Unsupervised Deep Event Stereo for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022.
  • SIFNet: Free-Form Image Inpainting Using a Color Split-Inpaint-Fuse Approach. Computer Vision and Image Understanding, 2022.
  • Global and Local Attention-Based Free-Form Image Inpainting. Sensors, 2020.
  • Deep Event Stereo Leveraged by Event-to-Image Translation. AAAI Conference on Artificial Intelligence (AAAI-21), 2021.

Challenge participations

  • Inverse Tone Mapping (AIM 2025, ICCV). Led the DITM entry, 5th of 67. Code
  • Image Super-Resolution x4 (NTIRE 2024, CVPR). 11th of 50, co-authored the challenge report.
  • Monocular Depth Estimation (2nd MDEC, CVPR 2023). 7th of 101.
  • Reversed ISP (AIM 2022, ECCV). 11th of 157.
  • Image Extreme Inpainting (AIM 2020, ECCV). 8th of 88.

Selected industrial projects

  • Monocular depth estimation for in-cabin monitoring (SL / Mobase / Mobis with Deep In Sight, 2024-2026)
  • 3D body keypoint, seat pose, gaze, and child presence estimation for in-cabin units (SL / Mobase with Deep In Sight, 2024-2026)
  • Pedestrian detection, distance, and predictive trajectory estimation for a Level 4 ADAS PoC (KADIF with DeltaX, 2023-2027)
  • Real-time eye distance estimation from a 3D light field display (Hyundai Mopic with DeltaX, 2023)
  • Worker detection, localization, and tracking with a long-wave infrared camera (Korea Railway Research Institute with DeltaX, 2022-2023)
  • Deep stereo matching and motion deblurring with a stereo event camera (SK hynix with CVIP Lab, 2021-2022)
  • High-speed pupil tracking for a holographic display (ETRI with CVIP Lab, 2019-2020)

Experience

Deep In Sight · Senior AI/ML Researcher, R&D (Nov 2024 - Present)

  • Driver and in-cabin monitoring: detection, segmentation, monocular depth, model compression, and pipeline integration.

DeltaX.ai · AI Researcher, Project Lead, Group Lead, Automotive Perception (Oct 2022 - Nov 2024)

  • Led work across driver and occupancy monitoring, depth and 3D, sensor fusion, and the delivery of several concurrent projects and PoCs.

Gachon University, CVIP Lab · Vision Researcher (Mar 2019 - Sep 2022)

  • Depth estimation, 3D reconstruction, and generative vision modeling.

Apex DMIT Ltd. · Business Development Analyst (Jul 2018 - Jan 2019)

University of Liberal Arts Bangladesh, EEE · Teaching Assistant (Feb 2018 - Jun 2018)

Education

Master of Engineering, IT Convergence Engineering · Gachon University (2019 - 2021)

  • Thesis on deep learning based image inpainting for irregular masks using attention.

Bachelor of Engineering, Electronics and Telecommunication Engineering · University of Liberal Arts Bangladesh (2013 - 2017)

Skills

Languages: Python, C++

Frameworks: PyTorch, TensorFlow, Keras, ONNX, TFLite, ROS

Libraries: NumPy, SciPy, OpenCV, scikit-learn, Pandas, Open3D

Tooling: Git, Docker, Jira, Confluence, Azure DevOps, Notion

Contact

Pinned Loading

  1. Global-and-Local-Attention-Based-Free-Form-Image-Inpainting Global-and-Local-Attention-Based-Free-Form-Image-Inpainting Public

    Official implementation of "Global and local attention-based free-form image inpainting"

    Python 61 8

  2. Inpainting-Evaluation-Metrics Inpainting-Evaluation-Metrics Public

    The goal of this repo is to provide a common evaluation script for image inpainting tasks. It contains some commonly used image quality metrics for inpainting (e.g., L1, L2, SSIM, PSNR and LPIPS).

    Python 9 3

  3. Image-Quality-Evaluation-Metrics Image-Quality-Evaluation-Metrics Public

    Implementation of Common Image Evaluation Metrics by Sayed Nadim (sayednadim.github.io). The repo is built based on full reference image quality metrics such as L1, L2, PSNR, SSIM, LPIPS. and featu…

    Python 27 4